For more than 14 years we have been integrating Mycobacterium tuberculosis genotyping with standard disease control information to examine the epidemiology of tuberculosis in San Francisco. Concurrently, we have used the natural laboratory provided by San Francisco to examine basic biological properties of the organism and to relate these properties to epidemiological features of tuberculosis. In this application we describe a set of studies designed to pursue observations made during the previous grant period suggesting that specific geographic lineages of M. tuberculosis have adapted to genetic differences in human populations. This conceptual framework provides a new perspective on the relationship between M. tuberculosis and humans. The association that we have observed has several potential explanations that fall into two main categories, sociological (epidemiological) and biological. In this application we will use a prospective study design to examine the epidemiological factors that possibly underlie the association in much greater detail than was possible retrospectively. Taken in context, the studies described herein comprise the second phase of a logical progression of investigations, the first phase of which was our analysis of retrospective data. The studies described in this application will utilize a more extensive examination of data from tuberculosis cases and their contacts to determine the extent to which epidemiological factors can be inferred to govern the transmission of M. tuberculosis to hosts of different race/ethnicities. If the results of the proposed studies suggest that the association is not entirely based on epidemiological factors, it would then be logical to begin a search for biological mechanisms. Thus, the third phase of the investigation would be an examination of racially/ethnically-associated differences in the human immune system. Specific examination of plausible immunological bases for the association, if the observation holds up under much closer scrutiny, would be conducted by our collaborators under separate funding. Confirming or refuting the concept that genetic variations in lineages interact in important ways with the genetic machinery of different human populations will have important implications for the design and testing of new vaccines and perhaps for clinical trials of drugs for treatment and prevention of tuberculosis.